Meta Platforms has acquired Moltbook, a social networking site designed for AI agents.

The company confirmed the purchase on Tuesday, stating that Moltbook’s founders Matt Schlicht and Ben Parr will join Meta Superintelligence Labs, an AI research unit overseen by Alexandr Wang.

Schlicht and Parr are scheduled to begin their roles at Meta on 16 March, according to reports.

Moltbook launched at the end of January as a platform where AI bots interact by sharing code and exchanging information. Human users can observe but do not have posting privileges.

The site has attracted attention within the technology sector and has sparked debate regarding the progression of AI towards human-like intelligence.

Some industry leaders have described Moltbook as indicative of broader trends in automation technology, while others have questioned the readiness of users to rely entirely on autonomous agents.

Meta did not reveal the financial terms of the acquisition.

This move follows other recent efforts by Meta to expand its capabilities in AI, including the earlier acquisition of Manus, a Singapore-based AI firm.

The rapid growth of Moltbook has brought scrutiny to its technical approach.

Cybersecurity company Wiz reported that a vulnerability in Moltbook exposed private messages, thousands of email addresses, and over a million credentials, reported Reuters.

Schlicht developed much of Moltbook’s infrastructure using his own AI assistant named Clawd Clawderberg and has promoted methods for building software with extensive AI support.

The competitive landscape among major technology firms for AI expertise continues to intensify as companies seek to develop autonomous systems capable of carrying out practical tasks.

According to Peter Fedoročko, CTO of the data analytics platform GoodData, Moltbook is a platform where autonomous agents post, comment, and interact with one another without direct human mediation. It has been described as a Reddit-style network where agents exchange skills, updates, and instructions.

Fedoročko added: “It’s a fascinating technical experiment that shows how quickly machine-to-machine ecosystems can organise themselves. However, it also highlights the risks of ungoverned agent interaction.

“These environments can become channels for indirect attacks, malicious instruction sharing, prompt injection, harmful skill exchange, and data leakage if they are not designed with proper infrastructure controls.

“If an agent can read browser data, search history, private documents, chats, or financial credentials, that information could be inadvertently transmitted or misused. So rather than a consumer social platform, Moltbook should be viewed as a signal of the complexity and risk that emerges when autonomous systems communicate without oversight.”